Tobacco smoke is a very complex sample, containing upwards of 4800 individual substances (Hoffmann, Hoffmann et al. 2001, Baker, Bishop 2004, Thielena, Klusb et al. 2008). The volatility of these substances vanes over a large range as does the concentration. Some of the substances are additives to the tobacco, but most are products of pyrolysis/combustion of natural components of tobacco, e.g. cellulose. The relatively high temperature (~ 950 C) allows complex reactions to occur, generating a very diverse collection of compounds. However, almost two-thirds of additives undergo <5% decomposition or reaction (Baker, Bishop 2004). The compounds are distributed between the gas phase are particle phase (liquid aerosol). For organic compounds, about 75% are in the particle phase and 25% in the gas phase. Sampling the particles is a very challenging task. Typically particles are collected on filters, extracted (and sometimes derivatized), and then analyzed by GC/MS or LC/MS (Figure 1). It has been known for some time that artifacts can be produced in this procedure. We have confirmed this using the setup shown in Figure 1. Research in our lab has recently been focused on developing new approaches to reduce or eliminate collection/extraction steps in the analysis of smoke, to minimize artifact formation.